Threshold scores

Applying to College
1

<p>Usually there's a preconceived "threshold" score that you have to obtain in a college if you are ever to have a fighting chance for acceptance. Probably for most ivies, the "threshold" is about 2100 (I don't know if this is too lenient, but if it is, just play along).</p>

<p>That being said, after that threshold is crossed, does it really matter if you score 2150 vs. scoring a 2400? I mean, after you cross it, does a higher score still significantly increase your chances of being accepted (AP statistics word "significantly" for those of you taking it), or is it more like anyone who crossed it is fair game? Because I didn't believe in the rumor that not even a 2400/36/4.0 will get you in some colleges, until I saw the Stanford threads.</p>

2

<p>This post from the MIT admissions blog addresses that question. The numbers are a little dated by now, but I believe the concept still applies.</p>

<p>[What’s</a> the big deal about 40^2? | MIT Admissions](<a href=“http://mitadmissions.org/blogs/entry/whats_the_big_deal_about_402]What’s”>What’s the big deal about 40^2? | MIT Admissions)</p>

3

<p>That post was good to see. It’s nice to know that we’re regarded as people and not numbers in the admissions process.</p>

4

<p>Brown University statistics:</p>

<p>Percentage of applicants w/ 800 on SAT CR accepted-20.6%
% accepted with 700-740 on SAT CR-10.5</p>

<p>Percentage of applicants w/ 800 on SAT Math accepted-17.1
% accepted with 700-740 on SAT math accepted-9.9</p>

<p>Percentage of applicants w/ 36 ACT accepted-27.8
Percentage of applicants w/ ACT of 33-35 accepted-11.2</p>

<p>Conclusion: Test scores matter quite a bit</p>

5

<p>From what I understand, MIT is the most open about not caring about test scores. It does seem to matter for most other schools, however.</p>

6

<p>Oh…yeah lol I just looked up stats for princeton. pretty much the same thing. I guess that blog only applied to MIT.</p>

7

<p>You will find the same sort of stats if you look at MIT too. I remember reading a different blog post that explained that even though people with, say, 800 on the math section had a higher acceptance rate, it wasn’t because they value an 800 over a 750. As in, they accepted the kid who qualified for USAMO over the kid who didn’t, and the same math strength that helped the first kid in math competitions also helped him get a higher SAT. More AP Stats: correlation doesn’t imply causation. :)</p>

8

<p>touche lol. so are you saying that there has to be some more academic success behind that higher score?</p>

9

<p>There exists bias in the readers. It would be natural to subconsciously favor the 2400 versus the 2100, all other things being equal.</p>

<p>I know this exists because in an off-the-record chat w/my HYP regional rep, he told me how he was advocating a grouping of applicants’ scores where, for instance, 2350-2400 got a red dot, 2250-2340 got a blue dot, etc. (these ranges are for example sake – only hypothetical). After the “dots” were assigned, the actual score would be hidden from readers. He said that the neglible diff between a 2400 and a 2360 scorer should not factor into any reader’s bias but his office knew it existed.</p>

<p>I don’t know if they adopted this system or not but its very discussion points to reader bias in favor of higher scores, regardless of whatever thresholds existed.</p>